Description of Problem: GinGin anaconda created boot disk does not boot. Everything is normal up intil kernel should start reading RAMDISK where it stops with no errors displayed. The test system has had linux running from as early as 7.2 time without any problems with bootdisks.
Looks like this is not anaconda problem, but kernel/mkinitrd problem. If there is no scsi card in system, boot disk creation is OK. With scsi module, initrd and kernel are too big for 1.44 floppy! [root@leon root]# cat /etc/modules.conf | grep scsi_hostadapter alias scsi_hostadapter aic7xxx [root@leon root]# mkbootdisk 2.4.19-0.pp.15 Insert a disk in /dev/fd0. Any information on the disk will be lost. Press <Enter> to continue or ^C to abort: gzip: stdout: No space left on device [root@leon root]# If I remove scsi_hostadapter line from /etc/modules.conf, boot disk creation is OK.
Works in beta2
Time tracking values updated
Boot disk (e100 and aic7xxx) does not fit on a floppy again. Anaconda does not tell user that there is not enough space. Message is posted on one of the virtual consoles. I tried to format disk using fdformat /dev/fd0u1760 and run mkbootdisk but resulting image would not boot.
Added code to verify that the initrd we create is the same size to avoid problems
Will the new initrd+kernel fit into regular 1.44 floppy or one needs to format 1.6 floppy? After I formatted /dev/fd0u1660, mkbootdisk created working bootdisk.
Anaconda now clearly detects out of space conditions. What about attempting to format /dev/fd0u1660 and try to recreate boot disk if regular 1.44 is not enough? This works for me.
*** This bug has been marked as a duplicate of 80605 ***
Changed to 'CLOSED' state since 'RESOLVED' has been deprecated.